# simnetpy
Python Package for the creation and analysis of similarity networks.
<!-- Project Organization
--------------------
.
├── AUTHORS.md
├── LICENSE
├── README.md
├── notebooks
└── src/simnet
├── clustering
├── datasets
├── graph
├── plotting
├── similarity
└── utils
-->
## Installation
Install latest release from PyPi with pip
```
pip install simnetpy
```
### Install from source
The source code for this project can be installed using `git` and `pip`.
Clone repository
```
git clone https://github.com/amarnane/simnetpy.git
cd simnetpy
pip install .
```
To remove the package simply use
```
pip uninstall simnetpy
```
### Developer Mode
To install in developer mode (have changes in source code update without reinstallation) add `-e` flag
```
pip install -e .
```
Note: removing the package is slightly more complicated and a different command is needed to uninstall
```
python setup.py develop -u
```
### Graph Tool
There is one dependency that cannot be installed through pip - `Graph-Tool`. This is a result of it's underlying `c++` dependencies.
The simplest method for python users is to make use of a conda environment, install this package using the commands above and install `graph-tool` using `conda-forge`
```
conda install -c conda-forge graph-tool
```
Note: this will not work on Windows. Alternative (conda independent) solutions can be found on the [Graph Tool Website](https://git.skewed.de/count0/graph-tool/-/wikis/installation-instructions#installing-via-conda)
# Using `simnetpy`
```Python
import simnet as sn
import numpy as np
# create mixed guassian data with 100 nodes, 2 dimensions and 3 equally sized clusters.
N = 100
sizes=np.array([34,33,33])
d = 2
dataset = sn.datasets.mixed_multi_guassian(len(sizes), d, N, sizes=sizes)
# calculate pairwise similarity
S = sn.pairwise_sim(dataset.X, metric='euclidean', norm=True)
# Create igraph Igraph from matrix
gg = sn.network_from_sim_mat(S, method='knn', K=10)
# print graph stats
print(gg.graph_stats())
# true cluster quality
cqual_ytrue = sn.clustering.cluster_quality(gg, dataset.y)
print(cqual_ytrue)
# cluster
ylabels = sn.clustering.spectral_clustering(gg, laplacian='lrw')
# cluster accuracy
cacc = sn.clustering.cluster_accuracy(dataset.y, ylabels)
print(cacc)
# predicted cluster quality
cqual = sn.clustering.cluster_quality(gg, ylabels)
print(cqual)
```
Raw data
{
"_id": null,
"home_page": "https://github.com/amarnane/simnetpy",
"name": "simnetpy",
"maintainer": "",
"docs_url": null,
"requires_python": ">=3.8",
"maintainer_email": "",
"keywords": "similarity-network network-construction networks graphs clustering graph-clustering community-detection",
"author": "Aidan Marnane",
"author_email": "aidan.marnane@gmail.com",
"download_url": "https://files.pythonhosted.org/packages/5a/c9/5de7b13517bd8b15682901630c8a6f9681b659410e46215ba8e321ee0f59/simnetpy-0.2.1.tar.gz",
"platform": null,
"description": "# simnetpy\n\nPython Package for the creation and analysis of similarity networks.\n\n<!-- Project Organization\n--------------------\n\n .\n \u251c\u2500\u2500 AUTHORS.md\n \u251c\u2500\u2500 LICENSE\n \u251c\u2500\u2500 README.md\n \u251c\u2500\u2500 notebooks\n \u2514\u2500\u2500 src/simnet\n \u251c\u2500\u2500 clustering\n \u251c\u2500\u2500 datasets\n \u251c\u2500\u2500 graph\n \u251c\u2500\u2500 plotting\n \u251c\u2500\u2500 similarity\n \u2514\u2500\u2500 utils \n-->\n\n\n## Installation\nInstall latest release from PyPi with pip\n```\npip install simnetpy\n```\n\n### Install from source\nThe source code for this project can be installed using `git` and `pip`. \nClone repository \n```\ngit clone https://github.com/amarnane/simnetpy.git\ncd simnetpy\npip install .\n```\n\nTo remove the package simply use \n```\npip uninstall simnetpy\n```\n\n### Developer Mode\nTo install in developer mode (have changes in source code update without reinstallation) add `-e` flag\n```\npip install -e .\n```\nNote: removing the package is slightly more complicated and a different command is needed to uninstall \n```\npython setup.py develop -u\n```\n\n### Graph Tool\nThere is one dependency that cannot be installed through pip - `Graph-Tool`. This is a result of it's underlying `c++` dependencies.\nThe simplest method for python users is to make use of a conda environment, install this package using the commands above and install `graph-tool` using `conda-forge`\n```\nconda install -c conda-forge graph-tool\n```\nNote: this will not work on Windows. Alternative (conda independent) solutions can be found on the [Graph Tool Website](https://git.skewed.de/count0/graph-tool/-/wikis/installation-instructions#installing-via-conda)\n\n# Using `simnetpy`\n```Python\nimport simnet as sn\nimport numpy as np\n\n# create mixed guassian data with 100 nodes, 2 dimensions and 3 equally sized clusters.\nN = 100\nsizes=np.array([34,33,33])\nd = 2\ndataset = sn.datasets.mixed_multi_guassian(len(sizes), d, N, sizes=sizes)\n\n# calculate pairwise similarity\nS = sn.pairwise_sim(dataset.X, metric='euclidean', norm=True)\n\n# Create igraph Igraph from matrix\ngg = sn.network_from_sim_mat(S, method='knn', K=10)\n\n# print graph stats\nprint(gg.graph_stats())\n\n# true cluster quality\ncqual_ytrue = sn.clustering.cluster_quality(gg, dataset.y)\nprint(cqual_ytrue)\n\n# cluster\nylabels = sn.clustering.spectral_clustering(gg, laplacian='lrw')\n\n# cluster accuracy\ncacc = sn.clustering.cluster_accuracy(dataset.y, ylabels)\nprint(cacc)\n\n# predicted cluster quality\ncqual = sn.clustering.cluster_quality(gg, ylabels)\nprint(cqual)\n```\n\n",
"bugtrack_url": null,
"license": "BSD",
"summary": "Python package for the Construction and Clustering of Similarity Networks",
"version": "0.2.1",
"project_urls": {
"Download": "https://github.com/amarnane/simnetpy",
"Homepage": "https://github.com/amarnane/simnetpy"
},
"split_keywords": [
"similarity-network",
"network-construction",
"networks",
"graphs",
"clustering",
"graph-clustering",
"community-detection"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "6910db92443b8f01b920754cd21a8ea9d4395b33a171cb79ce02843ecc5ffd2b",
"md5": "495934abc78f4e09d9b88fab5df225b5",
"sha256": "23db49bdd89c7e534b351c1f81884755e71d84512443bee70f345b6390478b5f"
},
"downloads": -1,
"filename": "simnetpy-0.2.1-py3-none-any.whl",
"has_sig": false,
"md5_digest": "495934abc78f4e09d9b88fab5df225b5",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.8",
"size": 53269,
"upload_time": "2023-11-02T13:37:59",
"upload_time_iso_8601": "2023-11-02T13:37:59.621747Z",
"url": "https://files.pythonhosted.org/packages/69/10/db92443b8f01b920754cd21a8ea9d4395b33a171cb79ce02843ecc5ffd2b/simnetpy-0.2.1-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "5ac95de7b13517bd8b15682901630c8a6f9681b659410e46215ba8e321ee0f59",
"md5": "e19c3c0cb99c9f34fc08086fb283d1ba",
"sha256": "fad2cc4d4ed621b13e0991c63b006850879cdb77aa56140f690ff954b2dc7448"
},
"downloads": -1,
"filename": "simnetpy-0.2.1.tar.gz",
"has_sig": false,
"md5_digest": "e19c3c0cb99c9f34fc08086fb283d1ba",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.8",
"size": 39572,
"upload_time": "2023-11-02T13:38:01",
"upload_time_iso_8601": "2023-11-02T13:38:01.242624Z",
"url": "https://files.pythonhosted.org/packages/5a/c9/5de7b13517bd8b15682901630c8a6f9681b659410e46215ba8e321ee0f59/simnetpy-0.2.1.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2023-11-02 13:38:01",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "amarnane",
"github_project": "simnetpy",
"travis_ci": false,
"coveralls": false,
"github_actions": false,
"lcname": "simnetpy"
}
JSON
